Rough — Granular neural network model for making treatment decisions of Hepatitis C

Hepatitis C virus is a massive health issue affecting significant portions of the world's population. Applying data preprocessing, feature reduction techniques, and generating rules based on the selected features for classification tasks are considered as important steps in the knowledge discovery in databases. This paper highlights a Rough-Granular Neural Networks model that incorporates Rough Sets and Artificial Neural Networks to make efficient data analysis and suggestive predictions. The Rough Sets is used as a powerful analysis tool for data pre-processing. It is used to reduce and choose the most relevant attributes for reducing the number of input vector to Artificial Neural Networks without reducing the basic knowledge of the information system. This is done to increase Classification accuracy of the proposed model. Resulting optimal data set is input to constructed neural network with supervised learning algorithm for classifying studied cases for testing new medication for HCV treatment. The experimental results show that the overall classification accuracy offered by the proposed model is a superlative result.